• Corpus ID: 9958665

Continuous Hindi Speech Recognition using Monophone based Acoustic Modeling

  title={Continuous Hindi Speech Recognition using Monophone based Acoustic Modeling},
  author={Ankith Jain Rakesh Kumar and Mohit Dua and Tripti Choudhary},
Speech is a natural way of communication and it provides an intuitive user interface to machines. Although the performance of automatic speech recognition (ASR) system is far from perfect. The overall performance of any speech recognition system is highly depends on the acoustic modeling. Hence generation of an accurate and robust acoustic model holds the key to satisfactory recognition performance. In this paper, we compare the performance of continuous Hindi speech recognition system with… 

Figures from this paper

Speaker Independent Speech Recognition Using Maximum Likelihood Approach for Isolated Words
A stochastic procedure for developing phoneme and word level acoustic models for Automatic Speech Recognition (ASR) system and compares and highlights the word and phoneme level acoustic model performances for Kannada language vocabulary.
Comparative Study of Isolated Word Recognition System for Hindi Language
This paper has described Comparative study Isolated Word Recognition System for Hindi Language using MFCC as feature extraction and KNN as pattern classification technique.
A Comprehensive View of Automatic Speech Recognition System - A Systematic Literature Review
  • Y. Kumar, Navdeep Singh
  • Computer Science
    2019 International Conference on Automation, Computational and Technology Management (ICACTM)
  • 2019
The recent advancement in the speech recognition system, robust method for the development of an automaticspeech recognition system and application of automatic speech Recognition system in different fields are elaborated.
Automatic Speech Recognition in Malayalam Using DNN-based Acoustic Modelling
DNN modelling shows significant improvement over GMM-HMM modelling framework in the design and also investigates the performance by varying number of hidden layers and hidden units.
HTK Based Speech Recognition Systems for Indian Regional languages: A Review
The aim of this paper is to give a brief overview of the Automatic Speech Recognition Systems which are built using HTK toolkit, and provides a literature survey of such systems which are build for recognizing Indian regional languages.
Feature Selection for Speech Recognition using Hidden Markov Model
An approach to the recognition of speech signal using frequency spectral information with Mel frequency for the improvement of speech feature representation in a HMM based recognition approach is presented.
Review on Different Approaches for Continuous Speech Recognition System
The basic idea ofspeech recognition, proposed types of speech recognition techniques, issues in speech recognition, different useful approaches for noise filtering, features extraction of the speech signal with its advantages and disadvantages and various pattern matching approaches for recognizing the speech of the speakers are presented.
Indian English speech recognition system for commonly used English Words using CMU tools
This paper is based on work done for the development of a speaker-independent Indian English speech recognition system and the system obtained best performance of 85.80 % for young (Eng2) model when trained using 16 GMMs (Gaussian Mixture Models).
Speech-to-Text Conversion in Indonesian Language Using a Deep Bidirectional Long Short-Term Memory Algorithm
The results showed that the deep bidirectional LSTM algorithm successfully converted speech to text in Indonesian, and MFCCs are more suitable than spectrograms as feature for speech-to-text conversion in Indonesian.
A Review on Automatic Speech Recognition System in Indian Regional Languages
Various types of techniques which is mostly used in automatic speech recognition are studied, which are used in web browser and spread sheets and other applications.


A speech recognition system for Hindi language using Hidden Markov Model Toolkit (HTK) to develop the system, which recognizes the isolated words using acoustic word model.
Using Gaussian Mixtures for Hindi Speech Recognition System
The statistical framework is reviewed and an iterative procedure to select an optimum number of Gaussian mixtures that exhibits maximum accuracy in the context of Hindi speech recognition system is presented.
Application of triphone clustering in acoustic modeling for continuous speech recognition in Bengali
The effect of triphone-based acoustic modeling over monophone based acoustic models in the context of continuous speech recognition in Bengali is looked at.
An Empirical Approach for Optimization of Acoustic Models in Hindi Speech Recognition Systems
The statistical framework is reviewed and an empirical approach to select the optimum number of Gaussian mixtures and appropriate degree of state tying in context of small amount of training data usually available for Indian languages, specifically Hindi is presented.
Acoustic Analysis for Automatic Speech Recognition
While the main focus in ASR is to obtain spectral envelope measures, human speech communication efficiently exploits the manipulation of one's vocal-cord vibration rate, and so F0 extraction and its integration into ASR are also reviewed.
Continuous density Hidden Markov Model for context dependent Hindi speech recognition
  • S. Sinha, S. Agrawal, Aruna Jain
  • Computer Science
    2013 International Conference on Advances in Computing, Communications and Informatics (ICACCI)
  • 2013
The authors estimate number of Gaussian mixture component for Hindi database based upon the size of vocabulary based on literature survey of research in this area.
Development of Application Specific Continuous Speech Recognition System in Hindi
The preliminary work towards a continuous speech recognition system in Hindi tailored to aid teaching Geometry in Primary schools is presented and results are presented.
Robust Features for Connected Hindi Digits Recognition
Connected digits recognition is important in many applications such as voice-dialing telephone, automated banking system, automatic data entry, PIN entry, etc. In this paper robust features such as
Large vocabulary speech recognition using subword units
An overview of decoding techniques for large vocabulary continuous speech recognition
  • X. Aubert
  • Computer Science
    Comput. Speech Lang.
  • 2002
A number of decoding strategies for large vocabulary continuous speech recognition (LVCSR) are examined from the viewpoint of their search space representation, and the main approaches are compared and some prospective views are formulated regarding possible future avenues.